The primary objective of this proposal is to demonstrate the feasibility and applicability of high-throughput profiling of alternate splice mRNA variants of five selected genes -CD44, CD46, ErbB, Muc1 and VEGF - in colon cancer detection and staging. The most common approach towards molecular profiling has been comparison of gene expression profiles using cDNA, PCR, tissue and oligonucleotide arrays. The complete sequence of the human genome has revealed approximately 35,000 genes, far less than the 100,000 or so proteins that account for the complexity of human biology. One explanation for this apparent anomaly is alternate splicing of mRNA. Alternate splicing of genes will have a profound impact in the discovery and validation of druggable targets and diagnostic biomarkers, e. g., in cancer (NCI).There is a dire need for high-throughput technologies capable of characterizing all mRNA variants within a biological sample. Profiling the relative variation in mRNA splice variants in tissues offers an additional level of resolution, complementing conventional gene expression analysis. This is an innovative new approach that promises to have the scalability of gene-based arrays combined with the sensitivity of proteomics-based molecular profiling. Jivan Biologics is developing a highly sensitive and scalable process, TransExpress TM that uniquely integrates microarray technology with computational biology and molecular biology techniques for discovery, full-length stitching and measurement of differential expression of alternately spliced forms of genes. The key personnel on this project will be Dr. Subha Sinivasan, Principal Investigator, and Jonathan Bingham, both of whom have extensive industry experience in genomics, microarray / gene chip technology and bioinfomatics. Jivan will also avail of consulting services from Dr. Manny Ares (UCSC) and Dr. Doug Black (UCLA), pre-eminent academic scientists in mRNA alternate splicing. The specific aims for SBIR Phase I are: (1) to determine the conditions to optimize Transexpress TMto profile mRNA splice variants in normal and neoplastic colon tissue samples, and (2) to obtain consensus mRNA variant profiles for normal and neoplastic colon tissues, by optimizing oligonucleotide probe selection. The success of this feasibility study will be depend on obtaining a significant reduction in the variation in intensities among constitutively expressed oligonucleotides for the same gene, and p-values of less than 0.01 for profiles from normal versus cancer colon tissues. The methodology used in this project comprises: (1) proprietary design, synthesis and spotting of aminelabeled oligonucleotides on glass slides, (2) hybridization of these microarrays with cy3-1abeled cRNA from 25 normal and 50 cancer colon tissue samples, and (3) proprietary computational analysis to stitch together and determine the relative expression levels of mRNA alternate splice variants in normal versus colon cancer tissues. These technologies will have both commercial and academic utility, consistent with the research interests of the Cancer Diagnosis Program (CDP) of the Division of Cancer Diagnosis and Treatment at the National Cancer Institutes (NCI). Upon successful completion of this project Jivan will use similar approach to a much larger set of genes and variants to create oligobased mRNA signatures for different stages of colon cancer as well as cancers in other sites.